create_podcasts.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Raspberry Pi Internet Radio podcast utility
# $Id: create_podcasts.py,v 1.2 2014/07/13 11:13:44 bob Exp $
#
# See Raspberry PI Radio constructors manual for instructions
#
# Author   : Bob Rathbone
# Web site : http://www.bobrathbone.com
#
#
# License: GNU V3, See https://www.gnu.org/copyleft/gpl.html
#
# Disclaimer: Software is provided as is and absolutly no warranties are implied or given.
#	     The authors shall not be liable for any loss or damage however caused.
#

import os
import sys
import urllib2
import unicodedata

from xml.dom.minidom import parseString

# Output errors to STDERR
stderr = sys.stderr.write;

# File locations
PlsDirectory = '/var/lib/mpd/playlists/'
RadioLibDir = '/var/lib/radiod/'
PodList = RadioLibDir + 'podcasts'

# MPD won't load greater than 350 entries in a playlist
MAX_ENTRIES = 250	# Maximum amount of entries in a playlist

# Execute system command
def execCommand(cmd):
	p = os.popen(cmd)
	return  p.readline().rstrip('\n')

# Create the output from the title, URL, length and file number
def createPlsOutput(title,url,length,filenumber):
        output = []
        output.append('File%s=%s' % (filenumber,url))
        output.append('Title%s=%s' % (filenumber,title))
        output.append('Length%s=%s' % (filenumber,length))
        return output

# Create the PLS or M3U file
def createPlsFile(basename,output,nlines):
        global duplicateCount
        uniqueCount = 1

	# Limit very large playlists
	if nlines > MAX_ENTRIES:
		nlines = MAX_ENTRIES

        if len(basename) < 1:
                filename = 'unknown'
	plsfile = basename + '.pls'
        filename = PlsDirectory + plsfile

        try:
                print 'Creating ' + filename 
                outfile = open(filename,'w')
                outfile.writelines("[playlist]\n")
                outfile.writelines("NumberOfEntries=%s\n"% nlines)
                outfile.writelines("Version=2\n")
		for item in output:
			outstr = item.encode('utf8', 'replace')
			outfile.write(outstr + "\n")
                outfile.close()
		# Load the new podcast into MPD
		print "Loading", plsfile + '\n'
		execCommand("mpc load " + plsfile)
        except:
                print "Failed to create",outfile
        return

# Execute system command
def execCommand(cmd):
	p = os.popen(cmd)
	return  p.readline().rstrip('\n')


def parseXml(url,filename,data,count):
	global errorCount
	global warningCount
	output = []

        try:
                dom = parseString(data)
        except Exception,e:
                print "Error:",e
                print "Error: Could not parse XML data from,", url + '\n'
                errorCount += 1
                return

        try:
		# Get all the items in the podcast
		items = dom.getElementsByTagName('item')
		filenumber = 0
		if count < 1:
			count = len(items)
		# Get the title and enclosure
		for node in items:
			filenumber += 1
			title = node.getElementsByTagName('title')[0].firstChild.nodeValue
			dtitle = title.encode('utf8', 'replace')
			print 'Title',filenumber,dtitle
			enclosure = node.getElementsByTagName('enclosure')[0]
			attributes = enclosure.attributes.items()
			url = dict(attributes).get('url') 
			length = dict(attributes).get('length') 
			type = dict(attributes).get('type') 
			durl = url.encode('utf8', 'replace')
			print durl,length
			output += createPlsOutput(title,url,length,filenumber)

			# Limit using count
			count -= 1
			if count < 1:
				break

		# Now create the file
		createPlsFile(filename,output,filenumber)

	except IndexError,e:
		print "Error:",e
		print "Error parsing", url
		errorCount += 1
		return "# DOM Error"

	return output

# Start of main routine
if os.getuid() != 0:
	print "This program can only be run as root user or using sudo"
	sys.exit(1) 

lineCount = 0	   # Line being processed (Including comments)
errorCount = 0	  # Errors
warningCount = 0	# Warnings
processedCount = 0      # Processed station count

# Set up playlist directory 
if not os.path.exists(PlsDirectory):
	execCommand ("mkdir -p " + PlsDirectory )

# Main processing loop
for line in open(PodList,'r'):
	lineCount += 1
	lines = []

	# Skip commented out or blank lines
	line = line.rstrip()    # Remove line feed
	if line[:1] == '#':
		continue

	# Ignore blank lines
	if len(line) < 1: 
		continue

	# Check start of title defined
	elif line[:1] != '[':
		stderr("Error: Missing left bracket [ in line %s in %s\n" % (lineCount,StationList))
		format()
		errorCount += 1
		continue

	processedCount += 1
	line = line.lstrip('[')

	# Get title and URL parts
	line = line.strip()
	lineparts = line.split(']')

	# Should be 2 parts (title and url)
	if len(lineparts) != 2:
		stderr("Error: Missing right bracket [ in line %s in %s\n" % (lineCount,StationList))
		format()
		errorCount += 1
		continue

	# Get title and URL from station definition
	title = lineparts[0].lstrip()
	print title
	# Extract number of feeds if present
	if ':' in title:
		titleparts = title.split(':')
		title = titleparts[0]
		count = int(titleparts[1])
	else:
		count = 0	# Means get all streams

	# Get the URL
	url = lineparts[1].lstrip()

	# Get the published URL and determine its type
	print 'Processing podcast ' + str(lineCount) + ': ' + url

	# Get the published URL to the stream file
	try:
		file = urllib2.urlopen(url)
		data = file.read()
		file.close()
	except:
		print "Error: Failed to retrieve ", url
		errorCount += 1
		continue

	# Create list from data
	filename = title.replace(' ', '_') 
	parseXml(url,filename,data,count)


# End of script